The present disclosure relates to a self-learning timer, and more specifically, to the application of a self-learning timer to traffic signals. Traffic signals are widely employed to regulate complex traffic patterns and provide assistance in locations where intersecting traffic patterns require driver interactions. Traffic signals generally act by directing some portion of the traffic to stop and wait, allowing an intersecting portion of traffic to safely proceed without collision. So-called “smart” traffic signals provide countdown timers for to notify travelers of how much time they have remaining in the signal's cycle to traverse a “go” signal, how long they can expect to wait at a “stop” signal, etc.